Opioid analgesics still rank as the most frequently prescribed drugs, with no broadly applicable alternative presently available. Opioid agonists, while creating analgesia in humans and other mammalian subjects, also have side effects. For example, opioid agonists are typically addictive, which can lead to widespread abuse, particularly for orally administered opioid analgesics such as oxycodone and hydromorphone. Addiction is characterized by tolerance, dependence, and drug craving. Another critical set of adverse effects includes respiratory depression (mostly CNS mediated), nausea, intestinal dysfunction, constipation, pruritus, bone loss, immune dysfunction, and more. The co-formulation—the subject of the present disclosure—seeks to reduce both addiction liability and peripheral adverse effects (e.g., constipation), requiring a unique set of required pharmacokinetic/pharmacological properties of the antagonist: neutral antagonist to avoid severe withdrawal, oral bioavailability, reduced access to the CNS, and a longer half-life than the agonist. In addition, oral bioavailability of the antagonist into the peripheral circulation is essential for reaching peripheral opioid receptors not residing in the gastrointestinal tract, affecting pruritus, bone loss, immune dysfunction, and more. Without considering all these qualities together, a co-formulation cannot be successful, or will at best be partially successful.
To address the addictive qualities of opioid agonists, the prior art teaches the use of co-formulations that include both opioid agonists and opioid antagonists. Most prior art co-formulations utilize opioid antagonists such as naltrexone or naloxone, which themselves have aversive effects in human and mammalian subjects. In fact, these compounds can trigger severe, and sometimes life-threatening, withdrawal symptoms. Such opioid antagonists are referred to in the present application as “aversive.” In addition, natrexone does not show peripheral selectivity, entering the brain with ease while naloxone has a very short half-life and poor bioavailability, characteristics that also disqualify these antagonists and their congener for the intended formulation. U.S. Pat. No. 6,627,635 to Palermo et al. and U.S. Pat. Nos. 6,696,066, 6,475,494 and U.S. Pat. No. 6,375,957 to Kaiko et al. provide examples of co-formulations that employ aversive opioid antagonists. Recent research by Marczak et al. emphasizes the aversive effects of both naltrexone and naloxone in opioid-dependent subjects. Marczak E., Jinsmaa Y., Li T., Bryant S. D., Tsuda Y., Okada Y., Lazaraus L. H. (Jul. 12, 2007) “[N-Allyl-Dmt1]-endomorphins are μ-opioid receptor antagonists lacking inverse agonist properties,” J. Pharmacol. Exp. Ther. Fast Forward. 
U.S. Pat. No. 6,713,488 B2 to Wolfgang Sadée et al., which is hereby incorporated herein by reference and is hereinafter referred to as “Sadée,” teaches the use of “neutral” opioid antagonists, which do not have aversive effects, to treat opioid agonist addiction. Sadée identifies as neutral antagonists certain naltrexone and naloxone analogs, including 6β-naltrexol and 6β-naltrexamide. Sadée also teaches that neutral antagonists can be used to treat other side-effects of opioid agonists, including constipation and respiratory depression.
Building on Sadée's research, United States Patent Application 2004/0024006 A1 to David Lew Simon, hereinafter referred to as “Simon,” proposes non-addictive opioid agonist/antagonist co-formulations that include primarily the aversive antagonist naltrexone while also incorporating neutral opioid antagonists such as 6β-naltrexol. Simon's disclosure is prophetic in nature, however, and lacks supporting, experimental data. Simon does not disclose anything about bioavailability of naltrexol. Certainly, the results described herein of the considerable selectivity of naltrexol in inhibiting the peripheral effects of an opioid agonist as compared to the central effects of that opioid agonist (hence, naltrexol has “peripheral selectivity”) are nowhere described in Simon. This lack of insight in Simon blocks the development of the co-formulation proposed here. For example, paragraphs [0107] and [108] of Simon discuss a known effective dosage of morphine (0.15) mg/kg body weight) and then suggest a completely unknown and untested range of possible doses for 6β-naltrexol from 0.00026-0.0015 mg/kg body weight, providing no data for such a formulation ever being effective. Similarly, in paragraph [0155] of Simon's application, Simon uses data provided by Kaiko for naltrexone (not naltrexol) in the U.S. Pat. No. 6,475,494 to propose that 0.5 to 12 mg of 6β-naltrexol be administered per 15 mg hydrocodone. These ratios of naltrexol to opioid range from 0.03:1 (i.e. 15.5-fold less naltrexol than opioid) to nearly a 1:1 ratio of naltrexol to opioid. Given the complete lack of supporting data, these two suggested dosage ranges spanning several orders of magnitude are uninformative and effectively meaningless, representing mere guesses. Further, Kaiko teaches the use of antagonists that have aversive effects in humans and precipitate severe withdrawal symptoms, lacking peripheral selectivity. Simon extrapolates to propose a dosage chart in paragraph [0156] for other kinds of opioid agonists, but also fails to consider the critical antagonist dosages needed to treat peripheral opioid adverse effects while providing no guidance on treating effectively addiction liability, which is CNS related. None of the prior art considers the dual peripheral and central effects of the co-formulation proposed here, requiring a unique set of properties.